Volume 13 | Issue 12 |

 Abstract

This paper presents an advanced approach for automated bone fracture detection using X-ray images combined with 2D-3D reconstruction and holographic visualization. The system enhances diagnostic accuracy by applying deep learning-based image processing to identify fracture regions with high precision. The extracted features are used to generate 3D anatomical structures that help clinicians understand fracture depth and alignment. These reconstructed models are projected as interactive holograms, enabling immersive medical visualization. The proposed method reduces diagnostic time, improves surgical planning, and supports remote healthcare applications. Experimental results demonstrate improved clarity, accuracy, and usability compared to traditional X-ray analysis.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

X-ray imaging, bone fracture detection, deep learning, image processing, 2D-3D reconstruction, holographic visualization, medical diagnostics, computer vision, healthcare technology, 3D modeling.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The rapid growth of IoT devices in smart city environments generated massive streams of real-time data, creating strong demand for efficient and secure anomaly detection mechanisms. In that vein, centrally administered AI methods are plagued by bandwidth limitations, latency, and privacy concerns, especially for large-scale deployments. This motivated the shift toward decentralized edge intelligence and federated learning for on-device analytics.Problem/Gap: Most of the anomaly detection frameworks used so far relied on cloud-centric architectures that could not cope with the demanding requirements imposed for real-time responses and did not give sufficient security to sensitive IoT data. Aim/Objective: The paper focuses on the design and evaluation of a federated learning-based edge intelligence framework that is able to conduct IoT anomaly detection tasks over smart city networks in real time. Methodology/Approach: The proposed framework combines Federated Averaging with adaptive edge-side aggregation to enable training of a distributed model without the sharing of raw sensor data. Experiments were conducted on the Edge-IIoTset dataset in an emulated edge computing environment using TensorFlow Federated with MQTT-based communication. Performance was compared against centralized machine learning and cloud-only approaches in different network and workload scenarios. The framework also used various communication optimization techniques to reduce synchronization overhead.Results / Findings: It improved the performance in anomaly detection by up to about 15% compared to the baseline, while comparatively reducing latency and communication costs since it allows for real-time, stable detection across the edge nodes. The system still maintained effective model convergence even with fluctuating network conditions. Implications / Significance: These results demonstrated that the proposed federated edge-based framework can bring scalability with the preservation of privacy and assurance of real-time analytics for IoT infrastructure in smart cities. Such a system will have a great impact on urban safety, predictive maintenance, and resilience of critical city services. Furthermore, this work also provided a practical basis for future deployments of autonomous distributed AI in smart environments.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Federated Learning; Edge Intelligence; IoT; Smart Cities; Anomaly Detection; Real-Time Systems; Privacy Preservation; Distributed Machine Learning

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The increasing incidence of hair loss and related scalp disorders has led to a surge in demand for natural hair care products. This study focused on the formulation and evaluation of a Polyherbal Hair Oil incorporating extracts from traditionally effective herbs like Bhringraj, Amla, Hibiscus, and Kalonji (Black Seed). The primary objective of this formulation was to stimulate hair growth, strengthen hair follicles, and address issues like dandruff. The prepared oil was thoroughly assessed for physicochemical parameters (pH, viscosity), organoleptic properties, and stability. The results confirmed the oil's safety, optimal physical attributes, and shelf life. Furthermore, evaluation of its antioxidant activity (via DPPH assay) and antimicrobial potential showed a significant inhibitory effect against fungal growth (dandruff-causing species) and high free radical scavenging capacity. In vivo studies demonstrated a notable improvement in both hair growth rate and density, establishing the formulation's efficacy and safety. This research concludes that the polyherbal oil offers a potent, natural, and effective alternative for hair care with a superior safety profile compared to synthetic products

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Keywords: Polyherbal, Hair Growth, Antioxidant, Formulation, Evaluation, Safety

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

QC and QA are vital parts of pharmaceutical production that assist in ensuring that every single remedy is safe, powerful, and of exceptional quality. QA generally avoids mistakes through proper structures, clean documentation, and adherence to GMP. However, QC includes the checking and checking out of raw materials, samples in manner, and finished products against predetermined criteria. Essentially, QA and QC characteristic in conjunction to hold consistency, lessen dangers, and take a look at that all drug treatments adhere to regulatory requirements throughout the producing system.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Quality Control, Quality Assurance, Good Manufacturing Practice, Documentation &Validation.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Cosmeceuticals are a fast-growing group of topical products that sit at the intersection of cosmetics and pharmaceuticals. They contain biologically active ingredients that provide therapeutic benefits to skin structure and function, going beyond just superficial enhancement. This shift from cosmetic-only to functionally active formulations shows the rising demand from consumers for skincare backed by scientific research. The global cosmeceutical market reached around USD 60 billion in 2023, with an expected compound annual growth rate of over 7%. It includes various bioactive categories like retinoids, vitamin C derivatives, hydroxy acids, peptides, growth factors, botanical antioxidants, niacinamide, and sunscreen actives. This review explores key cosmeceutical ingredient classes, detailing their molecular mechanisms, cellular targets, formulation science, clinical evidence of effectiveness, and dermatological uses in areas like photoaging, hyperpigmentation, acne vulgaris, rosacea, and barrier dysfunction. Retinoids impact gene transcription through retinoic acid receptors, boosting collagen production and normalizing the skin's outer layer. Ascorbic acid helps in collagen maturation while also acting as an antioxidant and reducing melanin production. Alpha-hydroxy acids increase skin cell turnover. Peptides activate fibroblasts and stimulate protein production. Careful formulation design, focusing on pH balance, light stability, vehicle choice, and penetration improvement, is essential for translating in vitro activity to clinical use. Regulatory standards differ worldwide. Cosmeceuticals often occupy unclear spaces between cosmetics, which must ensure safety without proving effectiveness, and drugs that need extensive clinical trials. However, strong evidence from randomized controlled trials confirms the effectiveness of many cosmeceutical groups for specific purposes. Future trends may include personalized formulas based on genetics that influence ingredient processing, designs compatible with the microbiome, advanced delivery methods, and the use of digital dermatology to assess outcomes objectively.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Cosmeceuticals , Retinoids , Vitamin C, Alpha-Hydroxy Acids, Hyperpigmentation, Dermatology, Skin Barrier

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

An innovative solution for modern agriculture, this project presents a comprehensive IoT-based smart polyhouse system for nurturing and monitoring crops. It addresses the challenges of traditional farming, such as manual monitoring, resource wastage, and vulnerability to climate change, by providing an automated and controlled environment for plant growth. The system uses various sensors to collect real-time data on critical parameters like soil moisture, temperature, and humidity. This information is processed by a central microcontroller and transmitted to a cloud platform, enabling farmers to monitor and control the polyhouse remotely via a user-friendly dashboard. By automating functions like irrigation, ventilation, and lighting based on predefined thresholds, the system ensures optimal growing conditions, leading to improved crop yield, resource conservation, and reduced labor. The design is a cost-effective, scalable, and sustainable approach to farming, promoting greater efficiency and resilience in food production.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Greenhouse, Polyhouse, Internet of Things (IoT), Automation, Remote Monitoring, Sensors, Microcontroller, Sustainable Agriculture, Smart Farming.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Web applications today are increasingly targeted by SQL Injection (SQLi) attacks, one of the most prevalent and damaging vulnerabilities affecting online systems. Traditional security mechanisms such as signature-based filters, firewalls, and manual validation provide incomplete protection against modern SQLi payloads that use obfuscation, encoding, and adversarial variations to bypass static rules. This work proposes a hybrid SQL Injection Detection System that integrates regular-expression-based preprocessing, deep-learning-driven feature extraction using a 1D Convolutional Neural Network (CNN), and ensemble classification using Random Forest. Multiple public datasets--including SQLiV3, CSIC HTTP dataset, and community payload repositories--are combined to construct a diverse and realistic dataset. Queries are tokenized, padded, and transformed into feature vectors, while handcrafted numerical features capture structural attributes of SQL inputs. The hybrid model performs real-time classification and distinguishes malicious queries from benign user inputs with high accuracy. Experimental results demonstrate improved detection performance compared to standalone ML or DL models. The system highlights the potential of lightweight hybrid architectures in strengthening application-layer security and mitigating SQLi threats in practical web environments. Index Terms-- SQL Injection Detection, Web Security, CNN Feature Extraction, Random Forest Classification, Hybrid Model, Input Validation, Machine Learning

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Index Terms-- SQL Injection Detection, Web Security, CNN Feature Extraction, Random Forest Classification, Hybrid Model, Input Validation, Machine Learning.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Bharat ki vividh murtikala shailiyon ka sangopang sinhavalokan

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Bharat ki vividh murtikala shailiyon ka sangopang sinhavalokan

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The concept of a fair trial is a cornerstone of the Indian legal system and an essential component of the rule of law. Rooted in constitutional guarantees under Articles 14, 20, 21, and 22, the right to a fair trial ensures justice, equality, due process, and protection against arbitrary State action. This analytical study examines the evolution, scope, and practical challenges of fair trial principles within India's judicial framework. It critically evaluates statutory provisions under the Code of Criminal Procedure, the Indian Evidence Act, and landmark judgments that have shaped contemporary understanding of due process. The study also explores systemic issues such as judicial delay, lack of legal awareness, unequal access to representation, and investigatory lapses that often obstruct the realization of fair trial rights. By assessing both doctrinal foundations and ground-level realities, the research aims to highlight gaps and offer insights for strengthening procedural fairness, transparency, and accountability in the administration of justice in India.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Fair trial,due process,arbitrary, accountability,fair procedure

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This study explores keylogger observation as a systematic approach to understanding real-time user behaviour on a computer system and identifying security-related events through detailed keystroke tracking. As digital activities increasingly move into sensitive areas like online banking, academic work, workplace communication, and personal information exchange, the need to detect unusual actions or hidden risks has become more important. By recording keystrokes in a controlled and authorized setup, the study examines how typing patterns, command sequences, and repeated user habits can reveal signs of unauthorized access, unsafe browsing tendencies, or potential system misuse. These observations not only support early detection of cyber threats but also help in analyzing human-computer interaction at a deeper level, offering valuable insights into how users think, respond, and navigate through digital environments. The collected data further contributes to improving user authentication models based on keystroke dynamics, enhancing monitoring techniques for educational and organizational systems, and assisting administrators in forming stronger policies for secure device usage. Overall, this work demonstrates that keylogger observation, when used responsibly and with proper ethical considerations, serves as a practical tool for strengthening system security, promoting safe digital behaviour, and improving the overall understanding of user activity in modern computing environments.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Keylogger; Keystroke Monitoring; Cybersecurity; User Activity Analysis; Ethical Surveillance; Data Logging; Computer Forensics; Parental Control; Behavioral Observation; System Security

  License

Creative Commons Attribution 4.0 and The Open Definition